[Users] meeting minutes for 2023-05-04

[Roland Haas]

Present: Roland, Peter, Zach, Gabriele, Leo, Sam

ET release timeline
===================

* release date will be May 24th
* Leo and Sam are sending emails to the contributors and drafting the
  release announcement
* inclusions are reviewed ok

kuibit
======

* Gabriele finds that installing kuibit using pip is getting harder and
  coming up with generic instructions is difficult
* Ubuntu 23.04 (and other new versions) prohibit use of `pip install
  --user` and strongly favor virtualenv
* Also link required package instructions as described in
https://bitbucket.org/einsteintoolkit/tickets/issues/2724/update-download-page

Testsuite status
================

* no updates
* test need to be re-run after fix to Fortran AliasedFunctions code
* Peter will re-run on sunrise once release branches are in place
* Roland has updated tests for Frontera and will re-run other tests
* Anuj will re-run tests on thornyflat

Gallery tests
=============

BBH: no update

BNS: Roland updated example

TOV: Peter will run

Multipatch: no update

Poisson: no volunteer, Leo and Sam to look for people

Gabriele suggested to have more detailed change list listing bugs more
explicitly
* Zach suggests using the script that shows changes for review by the
  release managers
* Gabriele suggests providing a link to this information, also to
  resolved tickets 
* link to revolved and closed tickets sorted by date:
https://bitbucket.org/einsteintoolkit/tickets/issues?status=resolved&status=invalid&status=closed&status=wontfix

questions
=========

* Gabriele mentioned large number of warning messages from Cactus when
  trying to run lean_public with presync_warn flags
https://bitbucket.org/einsteintoolkit/tickets/issues/2723/no-boundary-conditions-registered-for
** there was some more generic discussion about how presync handles it
   various options

* Leo brought up about HydroBaseX for CarpetX where currently there is
  no direct way to handle staggered variables 
** this is more generic problem with variables that may sensibly be
   defined or used with different sorts of staggering. Current HydroBase
   assumes a single global choice is good enough for all codes that
   interface and that translation between representations is almost
   error free.

emails
======

* existing emails by Shamim Haque, in conversation with Steve, ask
  Steve if is continuing the converstation

tickets
=======

*
  https://bitbucket.org/einsteintoolkit/tickets/issues/2722/getcomponents-places-kadaththorn-inside
  is needed for release, Roland will ask Steve

chair next week: Zach
minutes next week: Steve

Yours,
Roland
-- 
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://pgp.mit.edu .
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.einsteintoolkit.org/pipermail/users/attachments/20230504/b87217d8/attachment.sig>